Solvent-dependent properties of electrospun fibrous composites for bone tissue regeneration

Biodegradable polymer-ceramic composite scaffolds have gained importance in recent years in the field of orthopedic biomaterials and tissue engineering scaffolds for improving the rate of degradation and limited mechanical properties of bioactive ceramics. This study sought to create composites using the electrospinning process to achieve fibrous scaffolds with uniform fiber morphologies and uniform ceramic dispersions. Composites consisting of 20% hydroxyapatite/80% beta-tricalcium phosphate (20/80 HA/TCP) and poly (epsilon-caprolactone) (PCL) were fabricated. The 20/80 HA/TCP composition was chosen as the ceramic component because of previous reports of greater bone tissue formation in comparison with HA or TCP alone. For electrospinning, PCL was dissolved in either methylene chloride (Composite-MC) or a combination of methylene chloride (80%) and dimethylformamide (20%) (Composite-MC + DMF). Composite-MC mats contained a bimodal distribution of fiber diameters with nanofibers between larger, micron-sized fibers with an average pore size of 79.6 + or - 67 microm, whereas Composite-MC + DMF fibers had uniform fiber diameters with an average pore size of 7.0 + or - 4.2 microm. Elemental mapping determined that the ceramic was distributed throughout the mat and inside the fiber for both composites. However, physical characterization using differential scanning calorimetry (DSC) and mechanical testing revealed that the ceramic in the mats produced with MC + DMF were more uniformly dispersed than the ceramic in the mats produced with MC alone. Maximum tensile stress and strain were significantly higher for Composite-MC + DMF mats compared with Composite-MC mats and were comparable with the mechanical properties of mats of PCL alone. For both composites, there was molecular interaction between the PCL and the ceramic, as demonstrated by a maximum increase of approximately 10 degrees C in the glass transition values with the addition of the ceramic, as confirmed by Fourier transform infrared analysis. In addition, the crystallization behavior of the composites suggested that the ceramic was acting as a nucleating agent. Cell viability studies using human mesenchymal stem cells (MSC) showed that both composite scaffolds supported cell growth. However, cell numbers at early time points in culture were significantly higher on mats produced from MC + DMF compared with mats prepared with MC alone. Further examination revealed that cells were able to infiltrate the pores of the Composite-MC mats, but remained on the outer surface of the Composite-MC + DMF and unfilled PCL mats during the culture period. The results of this study demonstrate that the solvent or solvent combination used in preparing the electrospun composite mats plays a critical role in determining its properties, which may, in turn, affect cell behavior.